LangChain founder: a tiny one-practice-per-page wiki keeps agents from going stale
hwchase17 · x · 2026-11-22
Harrison Chase reshared the 'knowledge factory' idea: what makes AI work inside a company isn't better models or more context, but better knowledge of how the company works.
The quoted post details their internal practice: agents read a small wiki with one practice per page, 40 lines max; each page names its source file, which wins on conflict; dropped ideas get one line each so agents don't retry them. Chase adds that LangChain's internal GTM/content agents and LangSmith Engine all run similar learning loops.
More from coding & agent
- Scarlet: open-source AI video editor you control by voice, text, or drawing on frames — MatthewBerman · 2026-11-22
- AWS Marketplace adds x402 agentic payments so AI agents can buy APIs per request — kleffew94 · 2026-11-22
- Open-source Agent Skill hill-climbs local LLM serving speed, users report up to 5x gains — CatAstro_Piyush · 2026-11-22
- Walking a snow ridge in Genie 4, then rendering the same mountain four ways with MiniMax H3 in ComfyUI — PurzBeats · 2026-11-22
- Proof of Agent ecosystem booms: agents design their own join model, 24,815 onchain writes — adamamcbride · 2026-11-22
- Delvetown opens invites: humans and AI agents coexist in persistent community experiment — Miles_Brundage · 2026-11-22